docker-service-create (1) - Linux Man Pages

docker-service-create: Create a new service

NAME

docker-service-create - Create a new service

SYNOPSIS

docker service create [OPTIONS] IMAGE [COMMAND] [ARG...]

DESCRIPTION

Create a new service

OPTIONS

--config=
    Specify configurations to expose to the service

--constraint=
    Placement constraints

--container-label=
    Container labels

--credential-spec=
    Credential spec for managed service account (Windows only)

-d, --detach[=false]
    Exit immediately instead of waiting for the service to converge

--dns=
    Set custom DNS servers

--dns-option=
    Set DNS options

--dns-search=
    Set custom DNS search domains

--endpoint-mode="vip"
    Endpoint mode (vip or dnsrr)

--entrypoint=
    Overwrite the default ENTRYPOINT of the image

-e, --env=
    Set environment variables

--env-file=
    Read in a file of environment variables

--generic-resource=
    User defined resources

--group=
    Set one or more supplementary user groups for the container

--health-cmd=""
    Command to run to check health

--health-interval=
    Time between running the check (ms|s|m|h)

--health-retries=0
    Consecutive failures needed to report unhealthy

--health-start-period=
    Start period for the container to initialize before counting retries towards unstable (ms|s|m|h)

--health-timeout=
    Maximum time to allow one check to run (ms|s|m|h)

-h, --help[=false]
    help for create

--host=
    Set one or more custom host-to-IP mappings (host:ip)

--hostname=""
    Container hostname

--init[=false]
    Use an init inside each service container to forward signals and reap processes

--isolation=""
    Service container isolation mode

-l, --label=
    Service labels

--limit-cpu=
    Limit CPUs

--limit-memory=0
    Limit Memory

--log-driver=""
    Logging driver for service

--log-opt=
    Logging driver options

--mode="replicated"
    Service mode (replicated or global)

--mount=
    Attach a filesystem mount to the service

--name=""
    Service name

--network=
    Network attachments

--no-healthcheck[=false]
    Disable any container-specified HEALTHCHECK

--no-resolve-image[=false]
    Do not query the registry to resolve image digest and supported platforms

--placement-pref=
    Add a placement preference

-p, --publish=
    Publish a port as a node port

-q, --quiet[=false]
    Suppress progress output

--read-only[=false]
    Mount the container's root filesystem as read only

--replicas=
    Number of tasks

--replicas-max-per-node=0
    Maximum number of tasks per node (default 0 = unlimited)

--reserve-cpu=
    Reserve CPUs

--reserve-memory=0
    Reserve Memory

--restart-condition=""
    Restart when condition is met ("none"|"on-failure"|"any") (default "any")

--restart-delay=
    Delay between restart attempts (ns|us|ms|s|m|h) (default 5s)

--restart-max-attempts=
    Maximum number of restarts before giving up

--restart-window=
    Window used to evaluate the restart policy (ns|us|ms|s|m|h)

--rollback-delay=0s
    Delay between task rollbacks (ns|us|ms|s|m|h) (default 0s)

--rollback-failure-action=""
    Action on rollback failure ("pause"|"continue") (default "pause")

--rollback-max-failure-ratio=0
    Failure rate to tolerate during a rollback (default 0)

--rollback-monitor=0s
    Duration after each task rollback to monitor for failure (ns|us|ms|s|m|h) (default 5s)

--rollback-order=""
    Rollback order ("start-first"|"stop-first") (default "stop-first")

--rollback-parallelism=1
    Maximum number of tasks rolled back simultaneously (0 to roll back all at once)

--secret=
    Specify secrets to expose to the service

--stop-grace-period=
    Time to wait before force killing a container (ns|us|ms|s|m|h) (default 10s)

--stop-signal=""
    Signal to stop the container

--sysctl=
    Sysctl options

-t, --tty[=false]
    Allocate a pseudo-TTY

--update-delay=0s
    Delay between updates (ns|us|ms|s|m|h) (default 0s)

--update-failure-action=""
    Action on update failure ("pause"|"continue"|"rollback") (default "pause")

--update-max-failure-ratio=0
    Failure rate to tolerate during an update (default 0)

--update-monitor=0s
    Duration after each task update to monitor for failure (ns|us|ms|s|m|h) (default 5s)

--update-order=""
    Update order ("start-first"|"stop-first") (default "stop-first")

--update-parallelism=1
    Maximum number of tasks updated simultaneously (0 to update all at once)

-u, --user=""
    Username or UID (format: <name|uid>[:<group|gid>])

--with-registry-auth[=false]
    Send registry authentication details to swarm agents

-w, --workdir=""
    Working directory inside the container

SEE ALSO

docker-service(1)